利用收益预测与策略梯度两阶段众包评论集成

荣欢, 马廷淮

Two-Phase Crowdsourced Comment Integration Method Based on Reward Prediction and Policy Gradient

RONG Huan, MA Tinghuai

计算机科学与探索 . 2021, (8): 1476 -1489 . DOI: 10.3778/j.issn.1673-9418.2005023